Joint‐linkage mapping and GWAS reveal extensive genetic loci that regulate male inflorescence size in maize

Abstract

SummaryBoth insufficient and excessive male inflorescence size leads to a reduction in maize yield. Knowledge of the genetic architecture of male inflorescence is essential to achieve the optimum inflorescence size for maize breeding. In this study, we used approximately eight thousand inbreds, including both linkage populations and association populations, to dissect the genetic architecture of male inflorescence. The linkage populations include 25 families developed in the U.S. and 11 families developed in China. Each family contains approximately 200 recombinant inbred lines (RILs). The association populations include approximately 1000 diverse lines from the U.S. and China. All inbreds were genotyped by either sequencing or microarray. Inflorescence size was measured as the tassel primary branch number (TBN) and tassel length (TL). A total of 125 quantitative trait loci (QTLs) were identified (63 for TBN, 62 for TL) through linkage analyses. In addition, 965 quantitative trait nucleotides (QTNs) were identified through genomewide study (GWAS) at a bootstrap posterior probability (BPP) above a 5% threshold. These QTLs/QTNs include 24 known genes that were cloned using mutants, for example Ramosa3 (ra3), Thick tassel dwarf1 (td1), tasselseed2 (ts2), liguleless2 (lg2), ramosa1 (ra1), barren stalk1 (ba1), branch silkless1 (bd1) and tasselseed6 (ts6). The newly identified genes encode a zinc transporter (e.g. GRMZM5G838098 and GRMZM2G047762), the adapt in terminal region protein (e.g. GRMZM5G885628), O‐methyl‐transferase (e.g. GRMZM2G147491), helix‐loop‐helix (HLH) DNA‐binding proteins (e.g. GRMZM2G414252 and GRMZM2G042895) and an SBP‐box protein (e.g. GRMZM2G058588). These results provide extensive genetic information to dissect the genetic architecture of inflorescence size for the improvement of maize yield.

Complex genetic architecture underlies maize tassel domestication

“…To evaluate whether the QTLs identified in our maize‐teosinte population were also detected in maize populations, we compared the QTLs detected in this maize‐teosinte BC 2 S 3 population with the largest tassel QTL mapping study, which was recently performed by Wu et al . (), who used nearly 8000 maize RILs and inbred lines. Of the 25 TL and TBN QTLs mapped in this study, 14 (56%) QTLs overlapped with the QTLs reported by Wu et al .…”

Metabolic GWAS‐based dissection of genetic bases underlying the diversity of plant metabolism

“…In addition to the natural populations used in GWAS, well designed artificial populations have been developed to efficiently identify genetic variants determining complex traits. For example, a nested association mapping populations (NAM) of maize, derived from crosses between a reference line and other founder inbreds, have been used in GWAS of developmental traits and resistance to pathogens with improved mapping resolution (Kump et al ., ; Poland et al ., ; Tian et al ., ; Li et al ., ; Wu et al ., ). However, the extremely unbalanced parental compositions might dilute the GWAS efficiency (Xiao et al ., ).…”

Independent and Joint‐GWASfor growth traits inEucalyptusby assembling genome‐wide data for 3373 individuals across four breeding populations

“…Moreover, differently from crop breeding in which backcross introgression of high-value wild alleles into elite lines is commonplace, such a route is not an option in forest trees. An alternative strategy to detect high-value alleles by GWAS has been to develop more structured discovery populations, such as the nested association mapping (NAM) populations Wu et al, 2016). This approach puts the population through a one-generation bottleneck, raising some alleles to high and detectable frequency, while eliminating many others (Hamblin et al, 2011).…”

“…Due to the inherent challenges of creating large populations for GWAS in plants, most studies utilized populations smaller than a few hundred individuals, with the exception of studies using NAM populations in which several thousand individuals have been used (reviewed by Xiao et al, 2017). A potentially more viable alternative to gain statistical power in plants can be obtained by combining information from multiple populations using Meta-GWAS and Joint-GWAS (M€ agi & Morris, 2010;Yang et al, 2012;Bernal Rubio et al, 2016;Li et al, 2016;Wallace et al, 2016;Wu et al, 2016). Meta-GWAS combines the P-values from independent studies to increase the power to detect variants with small effect sizes and is a popular method for discovering new genetic risk variant in human datasets (Evangelou & Ioannidis, 2013).…”